The primary aim of this project is to understand how gene regulation generates precise spatial patterns in embryonic development. A major focus of developmental biology is to characterize how networks of genes and their products create distinct and spatially separated cell types. However, the chemical reactions and transport processes underlying pattern formation are subject to numerous sources of variability and noise. Extrinsic sources include variability in temperature, size and maternally-supplied factors. Intrinsic noise arises from the low concentrations of many biological molecules and the random aspects of cell shape, orientation and movement. For development to reliably form complex body plans, gene network dynamics must be robust to these disruptive influences. Investigating the generation and control of spatial noise requires a quantitative methodology. Choosing one of the genetically best characterized model systems for embryonic patterning, anterior-posterior segmentation in the fruit fly Drosophila, allows us to simplify the biological challenges, so that we can focus on noise characterization. Our ultimate goal, however, is to contribute to the understanding, and perhaps limiting, of human birth defects. Our work should also be directly relevant to the variable disease outcomes associated with incomplete gene penetrance and to error control mechanisms for limiting cancer. Since studies of noise and variability require careful quantitation, a major focus of our work is the development of robust image processing and statistical techniques for separating signal from different types of noise in whole embryo images. Variability between signals from different embryos provides data on the variability of global parameters;the different types of intrinsic noise provide data on within-embryo variation. We use modeling to understand how variability or noise arises in the segmentation gene network, and how they might be controlled. We model dynamics at the promoter level (using DNA structure) and at the network level (with simplified gene-gene interactions). At the fine-scale promoter level, fitting stochastic models to normal and experimentally perturbed expression data reveals the degree to which noise is generated in gene expression, as well as revealing noise-reducing mechanisms. Network level modeling of between-embryo variability data (normal and experimentally perturbed) allows us to test hypotheses on what interactions make segmentation patterns robust. Mathematical analysis is used to characterize the dynamics of these processes. We have 2 specific aims in characterizing the transition from maternal to zygotic control of noise and variability in segmentation: 1) Maternal gradient formation: we will focus on unanswered questions regarding the temporal and sub-cellular patterning of bicoid mRNA and protein. 2) Zygotic gene interactions: we will focus on how gap and pair-rule genes buffer against maternal noise and variability and increase spatial precision as segments are specified. . PUBLIC HEALTH RELEVANCE: Our project is aimed at understanding how genes interact in developing embryos to limit the effects of environmental and internal variability, so that tissues, organs and other parts of the body reliably form in the correct positions. Such gene regulation is also likely to be highly relevant to variable disease outcomes due to incomplete gene penetrance, and to error control mechanisms in cancer defense.